Qwen 2.5 1.5B: benchmark results
Alibaba's Qwen2.5-1.5B base model, pretrained with support for a 128K-token context (generating up to 8K tokens) and more than 29 languages, part of the 0.5B-72B Qwen2.5 family. Provider: Alibaba. Released 2024-09-19. Access: Open.
Aggregate Rating 1327 ± 13, rank #1431 of 1656 rated models, from 332 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LA Leaderboard - RagQuAS | 81.33 | Semantic Answer Similarity (x100) | 95.6 |
| LA Leaderboard - MGSM Catalan | 9.6 | Exact Match (%) | 90.8 |
| LA Leaderboard - MGSM Galician | 8 | Exact Match (%) | 84.6 |
| LA Leaderboard - XQuAD Spanish | 72.55 | F1 (%) | 84.6 |
| LA Leaderboard - MGSM Spanish | 4.4 | Exact Match (%) | 82.3 |
| LA Leaderboard - WiC Basque | 50.14 | Accuracy (%) | 80.9 |
| LA Leaderboard - XQuAD Catalan | 66.89 | F1 (%) | 73.8 |
| LA Leaderboard - TELEIA | 47.62 | Accuracy (%) | 72.1 |
| LA Leaderboard - HumorQA | 58.82 | Accuracy (%) | 70.6 |
| LA Leaderboard - CatalanQA | 72.66 | F1 (%) | 69.2 |
| LA Leaderboard - Belebele Galician | 66.33 | Accuracy (%) | 69.1 |
| LA Leaderboard - ClinTreatES | 58.28 | Semantic Answer Similarity (x100) | 68.7 |
Interactive version: theaggregate.ai/model?slug=qwen-2-5-1-5b · How It Works · Data refreshed daily, snapshot 2026-10-20.